| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.8, |
| "eval_steps": 100, |
| "global_step": 3000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.5915625745430588, |
| "epoch": 0.016, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.8e-07, |
| "loss": 0.8466, |
| "mean_token_accuracy": 0.8017937403172255, |
| "num_tokens": 340696.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.5926362803205848, |
| "epoch": 0.032, |
| "grad_norm": 5.59375, |
| "learning_rate": 3.8e-07, |
| "loss": 0.8487, |
| "mean_token_accuracy": 0.8023959375917912, |
| "num_tokens": 675930.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.591606929153204, |
| "epoch": 0.048, |
| "grad_norm": 4.96875, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 0.8302, |
| "mean_token_accuracy": 0.8044086921960115, |
| "num_tokens": 1007956.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.6204556789249182, |
| "epoch": 0.064, |
| "grad_norm": 4.9375, |
| "learning_rate": 7.8e-07, |
| "loss": 0.8457, |
| "mean_token_accuracy": 0.7998181059956551, |
| "num_tokens": 1340560.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.6393004363402724, |
| "epoch": 0.08, |
| "grad_norm": 4.3125, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 0.8343, |
| "mean_token_accuracy": 0.8025688324123621, |
| "num_tokens": 1679432.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.6593567343428731, |
| "epoch": 0.096, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 0.8046, |
| "mean_token_accuracy": 0.8078926544636488, |
| "num_tokens": 2021570.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.7051705561578274, |
| "epoch": 0.112, |
| "grad_norm": 2.8125, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 0.8246, |
| "mean_token_accuracy": 0.8029463469982148, |
| "num_tokens": 2359297.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.754444376192987, |
| "epoch": 0.128, |
| "grad_norm": 2.4375, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 0.8435, |
| "mean_token_accuracy": 0.79781529083848, |
| "num_tokens": 2683168.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.7250820267945528, |
| "epoch": 0.144, |
| "grad_norm": 1.90625, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 0.8195, |
| "mean_token_accuracy": 0.8042717296630144, |
| "num_tokens": 3019426.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.7252413807436824, |
| "epoch": 0.16, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.98e-06, |
| "loss": 0.8097, |
| "mean_token_accuracy": 0.8047077178955078, |
| "num_tokens": 3354390.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 0.7109885907173157, |
| "eval_chemistry_loss": 0.804060161113739, |
| "eval_chemistry_mean_token_accuracy": 0.8113441648483276, |
| "eval_chemistry_num_tokens": 3354390.0, |
| "eval_chemistry_runtime": 34.8657, |
| "eval_chemistry_samples_per_second": 14.341, |
| "eval_chemistry_steps_per_second": 3.585, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 0.7321163377761841, |
| "eval_physics_loss": 0.8287621140480042, |
| "eval_physics_mean_token_accuracy": 0.8033607420921326, |
| "eval_physics_num_tokens": 3354390.0, |
| "eval_physics_runtime": 41.3727, |
| "eval_physics_samples_per_second": 12.085, |
| "eval_physics_steps_per_second": 3.021, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.7267085557803512, |
| "epoch": 0.176, |
| "grad_norm": 2.234375, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 0.8058, |
| "mean_token_accuracy": 0.8033798363059759, |
| "num_tokens": 3684374.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.7212623069062829, |
| "epoch": 0.192, |
| "grad_norm": 1.828125, |
| "learning_rate": 2.38e-06, |
| "loss": 0.7989, |
| "mean_token_accuracy": 0.8043727930635214, |
| "num_tokens": 4023915.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7022871913388371, |
| "epoch": 0.208, |
| "grad_norm": 2.1875, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 0.7758, |
| "mean_token_accuracy": 0.8102051325142383, |
| "num_tokens": 4352045.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.6965231081470847, |
| "epoch": 0.224, |
| "grad_norm": 1.8203125, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 0.7726, |
| "mean_token_accuracy": 0.8089764468371868, |
| "num_tokens": 4681371.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.6956626677885651, |
| "epoch": 0.24, |
| "grad_norm": 1.6640625, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 0.7608, |
| "mean_token_accuracy": 0.8104452796280384, |
| "num_tokens": 5021784.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.708347842283547, |
| "epoch": 0.256, |
| "grad_norm": 1.9453125, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 0.7636, |
| "mean_token_accuracy": 0.809263763204217, |
| "num_tokens": 5367308.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.7138571232557297, |
| "epoch": 0.272, |
| "grad_norm": 2.03125, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 0.7721, |
| "mean_token_accuracy": 0.8050564888864755, |
| "num_tokens": 5699101.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.7222439678385854, |
| "epoch": 0.288, |
| "grad_norm": 1.7109375, |
| "learning_rate": 3.58e-06, |
| "loss": 0.7766, |
| "mean_token_accuracy": 0.8063390862196684, |
| "num_tokens": 6022837.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7117253359407186, |
| "epoch": 0.304, |
| "grad_norm": 2.265625, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 0.7535, |
| "mean_token_accuracy": 0.809041840955615, |
| "num_tokens": 6343826.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.7200546801090241, |
| "epoch": 0.32, |
| "grad_norm": 1.7265625, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 0.7524, |
| "mean_token_accuracy": 0.8073045868426562, |
| "num_tokens": 6670164.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 0.711975371837616, |
| "eval_chemistry_loss": 0.7488377690315247, |
| "eval_chemistry_mean_token_accuracy": 0.8110983457565307, |
| "eval_chemistry_num_tokens": 6670164.0, |
| "eval_chemistry_runtime": 34.8588, |
| "eval_chemistry_samples_per_second": 14.344, |
| "eval_chemistry_steps_per_second": 3.586, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 0.7178810501098633, |
| "eval_physics_loss": 0.7585415244102478, |
| "eval_physics_mean_token_accuracy": 0.8081966361999512, |
| "eval_physics_num_tokens": 6670164.0, |
| "eval_physics_runtime": 41.7087, |
| "eval_physics_samples_per_second": 11.988, |
| "eval_physics_steps_per_second": 2.997, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7111945543438196, |
| "epoch": 0.336, |
| "grad_norm": 1.578125, |
| "learning_rate": 4.18e-06, |
| "loss": 0.7313, |
| "mean_token_accuracy": 0.8092666126787662, |
| "num_tokens": 7003404.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.6962820990011096, |
| "epoch": 0.352, |
| "grad_norm": 1.8828125, |
| "learning_rate": 4.38e-06, |
| "loss": 0.712, |
| "mean_token_accuracy": 0.8133333619683981, |
| "num_tokens": 7342593.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.6933605512604117, |
| "epoch": 0.368, |
| "grad_norm": 1.6484375, |
| "learning_rate": 4.58e-06, |
| "loss": 0.6986, |
| "mean_token_accuracy": 0.8163042381405831, |
| "num_tokens": 7675569.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.7312716225162148, |
| "epoch": 0.384, |
| "grad_norm": 1.9296875, |
| "learning_rate": 4.78e-06, |
| "loss": 0.7319, |
| "mean_token_accuracy": 0.8038436233997345, |
| "num_tokens": 8012218.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.6996616214513779, |
| "epoch": 0.4, |
| "grad_norm": 1.796875, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 0.6917, |
| "mean_token_accuracy": 0.8134940605610609, |
| "num_tokens": 8350120.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.6945684082806111, |
| "epoch": 0.416, |
| "grad_norm": 1.6953125, |
| "learning_rate": 5.18e-06, |
| "loss": 0.678, |
| "mean_token_accuracy": 0.8151624854654074, |
| "num_tokens": 8688773.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.7035746525973081, |
| "epoch": 0.432, |
| "grad_norm": 1.8359375, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 0.6888, |
| "mean_token_accuracy": 0.8118865180760622, |
| "num_tokens": 9022094.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.6850203935056924, |
| "epoch": 0.448, |
| "grad_norm": 1.6640625, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 0.6727, |
| "mean_token_accuracy": 0.8186012674123049, |
| "num_tokens": 9360608.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.6806540634483099, |
| "epoch": 0.464, |
| "grad_norm": 1.875, |
| "learning_rate": 5.78e-06, |
| "loss": 0.6592, |
| "mean_token_accuracy": 0.8200464811176061, |
| "num_tokens": 9688896.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.6852828785777092, |
| "epoch": 0.48, |
| "grad_norm": 1.828125, |
| "learning_rate": 5.98e-06, |
| "loss": 0.6672, |
| "mean_token_accuracy": 0.8180346965789795, |
| "num_tokens": 10025864.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 0.7274510788917542, |
| "eval_chemistry_loss": 0.7009354829788208, |
| "eval_chemistry_mean_token_accuracy": 0.8133535690307617, |
| "eval_chemistry_num_tokens": 10025864.0, |
| "eval_chemistry_runtime": 34.811, |
| "eval_chemistry_samples_per_second": 14.363, |
| "eval_chemistry_steps_per_second": 3.591, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 0.7100022671222687, |
| "eval_physics_loss": 0.7013525366783142, |
| "eval_physics_mean_token_accuracy": 0.813646514415741, |
| "eval_physics_num_tokens": 10025864.0, |
| "eval_physics_runtime": 41.4407, |
| "eval_physics_samples_per_second": 12.065, |
| "eval_physics_steps_per_second": 3.016, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7103753052651882, |
| "epoch": 0.496, |
| "grad_norm": 1.6171875, |
| "learning_rate": 6.18e-06, |
| "loss": 0.6847, |
| "mean_token_accuracy": 0.812765546515584, |
| "num_tokens": 10362573.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.6982777336612344, |
| "epoch": 0.512, |
| "grad_norm": 1.8125, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 0.6716, |
| "mean_token_accuracy": 0.8169939771294594, |
| "num_tokens": 10694440.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.708159969188273, |
| "epoch": 0.528, |
| "grad_norm": 1.7421875, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 0.693, |
| "mean_token_accuracy": 0.8125004164874554, |
| "num_tokens": 11022357.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.699538960121572, |
| "epoch": 0.544, |
| "grad_norm": 1.828125, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 0.6711, |
| "mean_token_accuracy": 0.8153361644595861, |
| "num_tokens": 11345911.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.6910958087071777, |
| "epoch": 0.56, |
| "grad_norm": 1.5078125, |
| "learning_rate": 6.98e-06, |
| "loss": 0.6659, |
| "mean_token_accuracy": 0.8177970543503761, |
| "num_tokens": 11687225.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.6805725138634443, |
| "epoch": 0.576, |
| "grad_norm": 1.6484375, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 0.6565, |
| "mean_token_accuracy": 0.8207161612808704, |
| "num_tokens": 12015720.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.6865102933719754, |
| "epoch": 0.592, |
| "grad_norm": 1.84375, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 0.6638, |
| "mean_token_accuracy": 0.8175321508198976, |
| "num_tokens": 12343393.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.6841516578570008, |
| "epoch": 0.608, |
| "grad_norm": 1.6328125, |
| "learning_rate": 7.58e-06, |
| "loss": 0.6582, |
| "mean_token_accuracy": 0.8192304316908121, |
| "num_tokens": 12671474.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.7005494212731719, |
| "epoch": 0.624, |
| "grad_norm": 1.6015625, |
| "learning_rate": 7.78e-06, |
| "loss": 0.6674, |
| "mean_token_accuracy": 0.8156571265310049, |
| "num_tokens": 12996077.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.6937378702685237, |
| "epoch": 0.64, |
| "grad_norm": 1.6171875, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 0.6701, |
| "mean_token_accuracy": 0.8157013915479183, |
| "num_tokens": 13325291.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.7147737612724304, |
| "eval_chemistry_loss": 0.692807137966156, |
| "eval_chemistry_mean_token_accuracy": 0.8137030324935913, |
| "eval_chemistry_num_tokens": 13325291.0, |
| "eval_chemistry_runtime": 34.8141, |
| "eval_chemistry_samples_per_second": 14.362, |
| "eval_chemistry_steps_per_second": 3.591, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.684398841381073, |
| "eval_physics_loss": 0.6776748895645142, |
| "eval_physics_mean_token_accuracy": 0.8169920797348023, |
| "eval_physics_num_tokens": 13325291.0, |
| "eval_physics_runtime": 41.4558, |
| "eval_physics_samples_per_second": 12.061, |
| "eval_physics_steps_per_second": 3.015, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.6792764134705067, |
| "epoch": 0.656, |
| "grad_norm": 1.625, |
| "learning_rate": 8.18e-06, |
| "loss": 0.6538, |
| "mean_token_accuracy": 0.8188512332737445, |
| "num_tokens": 13671434.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.6741176716983318, |
| "epoch": 0.672, |
| "grad_norm": 1.703125, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 0.6479, |
| "mean_token_accuracy": 0.8206427529454231, |
| "num_tokens": 13993668.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.6748962674289942, |
| "epoch": 0.688, |
| "grad_norm": 1.5625, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 0.6489, |
| "mean_token_accuracy": 0.821071095019579, |
| "num_tokens": 14334907.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.6882562624290586, |
| "epoch": 0.704, |
| "grad_norm": 1.765625, |
| "learning_rate": 8.78e-06, |
| "loss": 0.6631, |
| "mean_token_accuracy": 0.8166982632130384, |
| "num_tokens": 14670639.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.6710160901769996, |
| "epoch": 0.72, |
| "grad_norm": 1.828125, |
| "learning_rate": 8.98e-06, |
| "loss": 0.6451, |
| "mean_token_accuracy": 0.8207047082483768, |
| "num_tokens": 14987995.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.6616904048249126, |
| "epoch": 0.736, |
| "grad_norm": 1.65625, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 0.6374, |
| "mean_token_accuracy": 0.8214268710464239, |
| "num_tokens": 15324659.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.6631792033091187, |
| "epoch": 0.752, |
| "grad_norm": 1.6171875, |
| "learning_rate": 9.38e-06, |
| "loss": 0.6379, |
| "mean_token_accuracy": 0.8226557578891516, |
| "num_tokens": 15657342.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.6816092301160097, |
| "epoch": 0.768, |
| "grad_norm": 1.640625, |
| "learning_rate": 9.58e-06, |
| "loss": 0.6582, |
| "mean_token_accuracy": 0.8162681456655264, |
| "num_tokens": 15988993.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.6584033869206906, |
| "epoch": 0.784, |
| "grad_norm": 1.6484375, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 0.6307, |
| "mean_token_accuracy": 0.8244805630296469, |
| "num_tokens": 16333857.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.6675900464877487, |
| "epoch": 0.8, |
| "grad_norm": 1.625, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 0.6444, |
| "mean_token_accuracy": 0.8208535581827163, |
| "num_tokens": 16664690.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.7186227707862854, |
| "eval_chemistry_loss": 0.6891804337501526, |
| "eval_chemistry_mean_token_accuracy": 0.812873167514801, |
| "eval_chemistry_num_tokens": 16664690.0, |
| "eval_chemistry_runtime": 34.7308, |
| "eval_chemistry_samples_per_second": 14.396, |
| "eval_chemistry_steps_per_second": 3.599, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.6725036988258362, |
| "eval_physics_loss": 0.6597840785980225, |
| "eval_physics_mean_token_accuracy": 0.8193980932235718, |
| "eval_physics_num_tokens": 16664690.0, |
| "eval_physics_runtime": 41.0989, |
| "eval_physics_samples_per_second": 12.166, |
| "eval_physics_steps_per_second": 3.041, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.6641527567058801, |
| "epoch": 0.816, |
| "grad_norm": 1.5625, |
| "learning_rate": 1.018e-05, |
| "loss": 0.6389, |
| "mean_token_accuracy": 0.8205956391990185, |
| "num_tokens": 16989307.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.6729596307501197, |
| "epoch": 0.832, |
| "grad_norm": 1.5859375, |
| "learning_rate": 1.038e-05, |
| "loss": 0.6418, |
| "mean_token_accuracy": 0.8202375840395689, |
| "num_tokens": 17317524.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.6599426086992025, |
| "epoch": 0.848, |
| "grad_norm": 1.40625, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 0.6326, |
| "mean_token_accuracy": 0.8230770550668239, |
| "num_tokens": 17650424.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.66338025983423, |
| "epoch": 0.864, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 0.6412, |
| "mean_token_accuracy": 0.8213306616991758, |
| "num_tokens": 17985547.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.6687672117725014, |
| "epoch": 0.88, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 0.6426, |
| "mean_token_accuracy": 0.8197343621402979, |
| "num_tokens": 18315392.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.6246686290949583, |
| "epoch": 0.896, |
| "grad_norm": 1.375, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 0.6014, |
| "mean_token_accuracy": 0.8294213119894266, |
| "num_tokens": 18667124.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.6640157140791416, |
| "epoch": 0.912, |
| "grad_norm": 1.5, |
| "learning_rate": 1.138e-05, |
| "loss": 0.6371, |
| "mean_token_accuracy": 0.820187458768487, |
| "num_tokens": 19002186.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.6525983370840549, |
| "epoch": 0.928, |
| "grad_norm": 1.6640625, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 0.6304, |
| "mean_token_accuracy": 0.8224839419126511, |
| "num_tokens": 19335013.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.6459019472822547, |
| "epoch": 0.944, |
| "grad_norm": 1.6171875, |
| "learning_rate": 1.178e-05, |
| "loss": 0.6208, |
| "mean_token_accuracy": 0.8251514233648777, |
| "num_tokens": 19662733.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6452493501827121, |
| "epoch": 0.96, |
| "grad_norm": 1.5625, |
| "learning_rate": 1.198e-05, |
| "loss": 0.6276, |
| "mean_token_accuracy": 0.8237052172422409, |
| "num_tokens": 19996789.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 0.7299397034645081, |
| "eval_chemistry_loss": 0.6870628595352173, |
| "eval_chemistry_mean_token_accuracy": 0.8117328572273255, |
| "eval_chemistry_num_tokens": 19996789.0, |
| "eval_chemistry_runtime": 34.7851, |
| "eval_chemistry_samples_per_second": 14.374, |
| "eval_chemistry_steps_per_second": 3.593, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 0.6775157516002654, |
| "eval_physics_loss": 0.646020770072937, |
| "eval_physics_mean_token_accuracy": 0.8218063368797303, |
| "eval_physics_num_tokens": 19996789.0, |
| "eval_physics_runtime": 41.4104, |
| "eval_physics_samples_per_second": 12.074, |
| "eval_physics_steps_per_second": 3.019, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.6451076414436102, |
| "epoch": 0.976, |
| "grad_norm": 1.5859375, |
| "learning_rate": 1.218e-05, |
| "loss": 0.6221, |
| "mean_token_accuracy": 0.825892923027277, |
| "num_tokens": 20332636.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.6509507959708571, |
| "epoch": 0.992, |
| "grad_norm": 1.453125, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 0.6214, |
| "mean_token_accuracy": 0.8233620677143335, |
| "num_tokens": 20662209.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.6204982601106167, |
| "epoch": 1.008, |
| "grad_norm": 1.46875, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 0.5988, |
| "mean_token_accuracy": 0.8286421094089746, |
| "num_tokens": 21005731.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.6242268454283476, |
| "epoch": 1.024, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 0.6046, |
| "mean_token_accuracy": 0.82848214879632, |
| "num_tokens": 21335407.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.6163413025438785, |
| "epoch": 1.04, |
| "grad_norm": 1.5078125, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 0.598, |
| "mean_token_accuracy": 0.8299466107040644, |
| "num_tokens": 21668490.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.6118087394163012, |
| "epoch": 1.056, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 0.5883, |
| "mean_token_accuracy": 0.8311309419572354, |
| "num_tokens": 22006039.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.6269845755770802, |
| "epoch": 1.072, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 0.6043, |
| "mean_token_accuracy": 0.8276471871882677, |
| "num_tokens": 22348206.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.6239592241123318, |
| "epoch": 1.088, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 0.5947, |
| "mean_token_accuracy": 0.8277118816971779, |
| "num_tokens": 22676163.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.6154211649671197, |
| "epoch": 1.104, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.378e-05, |
| "loss": 0.5951, |
| "mean_token_accuracy": 0.8299165233969689, |
| "num_tokens": 23009279.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6186880182474852, |
| "epoch": 1.12, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.398e-05, |
| "loss": 0.5987, |
| "mean_token_accuracy": 0.827697791159153, |
| "num_tokens": 23346828.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 0.6806538386344909, |
| "eval_chemistry_loss": 0.6874776482582092, |
| "eval_chemistry_mean_token_accuracy": 0.8116622562408448, |
| "eval_chemistry_num_tokens": 23346828.0, |
| "eval_chemistry_runtime": 35.1659, |
| "eval_chemistry_samples_per_second": 14.218, |
| "eval_chemistry_steps_per_second": 3.555, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.6262999680042267, |
| "eval_physics_loss": 0.6376790404319763, |
| "eval_physics_mean_token_accuracy": 0.8227445411682129, |
| "eval_physics_num_tokens": 23346828.0, |
| "eval_physics_runtime": 41.445, |
| "eval_physics_samples_per_second": 12.064, |
| "eval_physics_steps_per_second": 3.016, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.5965470356866718, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.418e-05, |
| "loss": 0.5793, |
| "mean_token_accuracy": 0.8342778768390418, |
| "num_tokens": 23689161.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.6170177994295954, |
| "epoch": 1.152, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 0.5984, |
| "mean_token_accuracy": 0.8301392119377852, |
| "num_tokens": 24016439.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6231384797021746, |
| "epoch": 1.168, |
| "grad_norm": 1.640625, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 0.6024, |
| "mean_token_accuracy": 0.8265216089785099, |
| "num_tokens": 24344431.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6168253228068352, |
| "epoch": 1.184, |
| "grad_norm": 1.734375, |
| "learning_rate": 1.478e-05, |
| "loss": 0.5912, |
| "mean_token_accuracy": 0.8295746680349112, |
| "num_tokens": 24668087.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.6063494171947241, |
| "epoch": 1.2, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.498e-05, |
| "loss": 0.5805, |
| "mean_token_accuracy": 0.8319725230336189, |
| "num_tokens": 25000629.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6178384765982627, |
| "epoch": 1.216, |
| "grad_norm": 1.6796875, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 0.5998, |
| "mean_token_accuracy": 0.8278374817222357, |
| "num_tokens": 25324324.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6116716200485826, |
| "epoch": 1.232, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 0.5838, |
| "mean_token_accuracy": 0.8310758478939533, |
| "num_tokens": 25658795.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.6036085536703467, |
| "epoch": 1.248, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 0.5835, |
| "mean_token_accuracy": 0.8323761031031609, |
| "num_tokens": 25993177.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.6078255159780384, |
| "epoch": 1.264, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.578e-05, |
| "loss": 0.5819, |
| "mean_token_accuracy": 0.8314513519406319, |
| "num_tokens": 26329726.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.6271371256560088, |
| "epoch": 1.28, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 0.6049, |
| "mean_token_accuracy": 0.8257384181022644, |
| "num_tokens": 26664093.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 0.6834847524166107, |
| "eval_chemistry_loss": 0.6882474422454834, |
| "eval_chemistry_mean_token_accuracy": 0.8116786751747131, |
| "eval_chemistry_num_tokens": 26664093.0, |
| "eval_chemistry_runtime": 34.7916, |
| "eval_chemistry_samples_per_second": 14.371, |
| "eval_chemistry_steps_per_second": 3.593, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.6253828897476196, |
| "eval_physics_loss": 0.6271702647209167, |
| "eval_physics_mean_token_accuracy": 0.8248588938713074, |
| "eval_physics_num_tokens": 26664093.0, |
| "eval_physics_runtime": 41.5612, |
| "eval_physics_samples_per_second": 12.03, |
| "eval_physics_steps_per_second": 3.008, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.5998634301126003, |
| "epoch": 1.296, |
| "grad_norm": 1.5625, |
| "learning_rate": 1.618e-05, |
| "loss": 0.5787, |
| "mean_token_accuracy": 0.8333883356302977, |
| "num_tokens": 26994877.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6259857648983598, |
| "epoch": 1.312, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.638e-05, |
| "loss": 0.6016, |
| "mean_token_accuracy": 0.8277184132486581, |
| "num_tokens": 27328198.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.605760240368545, |
| "epoch": 1.328, |
| "grad_norm": 1.5546875, |
| "learning_rate": 1.658e-05, |
| "loss": 0.5891, |
| "mean_token_accuracy": 0.830071647465229, |
| "num_tokens": 27658818.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.6093621645122766, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 1.546875, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 0.5882, |
| "mean_token_accuracy": 0.8291759472340345, |
| "num_tokens": 27991108.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.6125000966712832, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.698e-05, |
| "loss": 0.582, |
| "mean_token_accuracy": 0.8292897183448076, |
| "num_tokens": 28327775.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6071364052593708, |
| "epoch": 1.376, |
| "grad_norm": 1.671875, |
| "learning_rate": 1.718e-05, |
| "loss": 0.5823, |
| "mean_token_accuracy": 0.8310487173497677, |
| "num_tokens": 28666116.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.5939128069207072, |
| "epoch": 1.392, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 0.5723, |
| "mean_token_accuracy": 0.8350477192550898, |
| "num_tokens": 29004293.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.6024157986044884, |
| "epoch": 1.408, |
| "grad_norm": 1.421875, |
| "learning_rate": 1.758e-05, |
| "loss": 0.5827, |
| "mean_token_accuracy": 0.8312420520931483, |
| "num_tokens": 29330131.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6242644298821688, |
| "epoch": 1.424, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.6039, |
| "mean_token_accuracy": 0.8263261545449495, |
| "num_tokens": 29661850.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.5897390013560653, |
| "epoch": 1.44, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.798e-05, |
| "loss": 0.5687, |
| "mean_token_accuracy": 0.8350921101868153, |
| "num_tokens": 29999392.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 0.6641502072811126, |
| "eval_chemistry_loss": 0.6906805634498596, |
| "eval_chemistry_mean_token_accuracy": 0.8092845277786255, |
| "eval_chemistry_num_tokens": 29999392.0, |
| "eval_chemistry_runtime": 34.7482, |
| "eval_chemistry_samples_per_second": 14.389, |
| "eval_chemistry_steps_per_second": 3.597, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.6082331819534301, |
| "eval_physics_loss": 0.6185559034347534, |
| "eval_physics_mean_token_accuracy": 0.8265736656188964, |
| "eval_physics_num_tokens": 29999392.0, |
| "eval_physics_runtime": 41.4301, |
| "eval_physics_samples_per_second": 12.069, |
| "eval_physics_steps_per_second": 3.017, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.5889361893758178, |
| "epoch": 1.456, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 0.5722, |
| "mean_token_accuracy": 0.8360798362642526, |
| "num_tokens": 30341004.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.6223346112295985, |
| "epoch": 1.472, |
| "grad_norm": 1.484375, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 0.599, |
| "mean_token_accuracy": 0.8281814876943827, |
| "num_tokens": 30667689.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.6086510250344872, |
| "epoch": 1.488, |
| "grad_norm": 1.25, |
| "learning_rate": 1.858e-05, |
| "loss": 0.5865, |
| "mean_token_accuracy": 0.8305441569536924, |
| "num_tokens": 30999697.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.5955386120826006, |
| "epoch": 1.504, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.878e-05, |
| "loss": 0.5699, |
| "mean_token_accuracy": 0.8333926767110824, |
| "num_tokens": 31332775.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.6133549332618713, |
| "epoch": 1.52, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.898e-05, |
| "loss": 0.5915, |
| "mean_token_accuracy": 0.8288876831531524, |
| "num_tokens": 31666521.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.6097657412290574, |
| "epoch": 1.536, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.918e-05, |
| "loss": 0.5866, |
| "mean_token_accuracy": 0.8288511786609888, |
| "num_tokens": 31998232.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.5852528993040323, |
| "epoch": 1.552, |
| "grad_norm": 1.453125, |
| "learning_rate": 1.938e-05, |
| "loss": 0.5684, |
| "mean_token_accuracy": 0.8353338003158569, |
| "num_tokens": 32328085.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.5982964089140296, |
| "epoch": 1.568, |
| "grad_norm": 1.375, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 0.5753, |
| "mean_token_accuracy": 0.832823446765542, |
| "num_tokens": 32656603.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.6145529938861728, |
| "epoch": 1.584, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.978e-05, |
| "loss": 0.5882, |
| "mean_token_accuracy": 0.8275394171476365, |
| "num_tokens": 32983769.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.6043317306786775, |
| "epoch": 1.6, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 0.5823, |
| "mean_token_accuracy": 0.8310152795165777, |
| "num_tokens": 33309282.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 0.6794673941135406, |
| "eval_chemistry_loss": 0.6896947026252747, |
| "eval_chemistry_mean_token_accuracy": 0.8101202611923217, |
| "eval_chemistry_num_tokens": 33309282.0, |
| "eval_chemistry_runtime": 34.8102, |
| "eval_chemistry_samples_per_second": 14.364, |
| "eval_chemistry_steps_per_second": 3.591, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.607559362411499, |
| "eval_physics_loss": 0.6121883392333984, |
| "eval_physics_mean_token_accuracy": 0.8272243642807007, |
| "eval_physics_num_tokens": 33309282.0, |
| "eval_physics_runtime": 41.0884, |
| "eval_physics_samples_per_second": 12.169, |
| "eval_physics_steps_per_second": 3.042, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.5851200748234987, |
| "epoch": 1.616, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 0.5675, |
| "mean_token_accuracy": 0.8361915960907936, |
| "num_tokens": 33647616.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.5935754748061299, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 0.5753, |
| "mean_token_accuracy": 0.8332207947969437, |
| "num_tokens": 33977586.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.5731598405167461, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 1.390625, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 0.5536, |
| "mean_token_accuracy": 0.8382600821554661, |
| "num_tokens": 34312436.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.5990395007655025, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 1.5390625, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 0.5747, |
| "mean_token_accuracy": 0.8332929655909538, |
| "num_tokens": 34641549.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.589907999895513, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 0.5685, |
| "mean_token_accuracy": 0.8342956956475973, |
| "num_tokens": 34977031.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.5947070028632879, |
| "epoch": 1.696, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 0.5736, |
| "mean_token_accuracy": 0.8329529393464327, |
| "num_tokens": 35300962.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.595324145630002, |
| "epoch": 1.712, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 0.574, |
| "mean_token_accuracy": 0.831612591072917, |
| "num_tokens": 35642662.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.581332896463573, |
| "epoch": 1.728, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 0.5648, |
| "mean_token_accuracy": 0.8354741148650646, |
| "num_tokens": 35980993.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.5898414205759763, |
| "epoch": 1.744, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 0.568, |
| "mean_token_accuracy": 0.834344669431448, |
| "num_tokens": 36326797.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.5798839410766959, |
| "epoch": 1.76, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.978e-05, |
| "loss": 0.5617, |
| "mean_token_accuracy": 0.8352703254669904, |
| "num_tokens": 36654797.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 0.662034518957138, |
| "eval_chemistry_loss": 0.6911905407905579, |
| "eval_chemistry_mean_token_accuracy": 0.8086481132507324, |
| "eval_chemistry_num_tokens": 36654797.0, |
| "eval_chemistry_runtime": 34.786, |
| "eval_chemistry_samples_per_second": 14.374, |
| "eval_chemistry_steps_per_second": 3.593, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.5805578632354736, |
| "eval_physics_loss": 0.6039645671844482, |
| "eval_physics_mean_token_accuracy": 0.8294073853492737, |
| "eval_physics_num_tokens": 36654797.0, |
| "eval_physics_runtime": 41.4109, |
| "eval_physics_samples_per_second": 12.074, |
| "eval_physics_steps_per_second": 3.019, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.6007018189877271, |
| "epoch": 1.776, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 0.5839, |
| "mean_token_accuracy": 0.8299636762589216, |
| "num_tokens": 36989310.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.5859053635969758, |
| "epoch": 1.792, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 0.5609, |
| "mean_token_accuracy": 0.8349329460412264, |
| "num_tokens": 37318535.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.5717526819556952, |
| "epoch": 1.808, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 0.5516, |
| "mean_token_accuracy": 0.837704149633646, |
| "num_tokens": 37658889.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.5855952553451061, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 1.25, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 0.5609, |
| "mean_token_accuracy": 0.8358533851802349, |
| "num_tokens": 37995198.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.576733535528183, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 1.2421875, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 0.5602, |
| "mean_token_accuracy": 0.8364770855754614, |
| "num_tokens": 38325542.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.5805189371109009, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 0.5562, |
| "mean_token_accuracy": 0.8367651741951704, |
| "num_tokens": 38668148.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.5755893625319004, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 0.5567, |
| "mean_token_accuracy": 0.8351109046489, |
| "num_tokens": 39001652.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.593781272508204, |
| "epoch": 1.888, |
| "grad_norm": 1.375, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 0.5698, |
| "mean_token_accuracy": 0.8326846692711115, |
| "num_tokens": 39333474.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.5997770514339209, |
| "epoch": 1.904, |
| "grad_norm": 1.25, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 0.5793, |
| "mean_token_accuracy": 0.8314935185015202, |
| "num_tokens": 39663055.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.5944939412176609, |
| "epoch": 1.92, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 0.5733, |
| "mean_token_accuracy": 0.8316057369112968, |
| "num_tokens": 39989824.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 0.6759118571281433, |
| "eval_chemistry_loss": 0.6883504986763, |
| "eval_chemistry_mean_token_accuracy": 0.8099100780487061, |
| "eval_chemistry_num_tokens": 39989824.0, |
| "eval_chemistry_runtime": 35.0777, |
| "eval_chemistry_samples_per_second": 14.254, |
| "eval_chemistry_steps_per_second": 3.564, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.5913804955482483, |
| "eval_physics_loss": 0.5959852337837219, |
| "eval_physics_mean_token_accuracy": 0.8307065281867981, |
| "eval_physics_num_tokens": 39989824.0, |
| "eval_physics_runtime": 41.4833, |
| "eval_physics_samples_per_second": 12.053, |
| "eval_physics_steps_per_second": 3.013, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.5832644954323769, |
| "epoch": 1.936, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 0.5641, |
| "mean_token_accuracy": 0.8353390946984292, |
| "num_tokens": 40325070.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.5566705971956253, |
| "epoch": 1.952, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 0.5374, |
| "mean_token_accuracy": 0.8413328900933266, |
| "num_tokens": 40659614.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.5830600494518876, |
| "epoch": 1.968, |
| "grad_norm": 1.15625, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 0.5625, |
| "mean_token_accuracy": 0.835172512754798, |
| "num_tokens": 40991508.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.581772717833519, |
| "epoch": 1.984, |
| "grad_norm": 1.25, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 0.5629, |
| "mean_token_accuracy": 0.8342908933758736, |
| "num_tokens": 41328779.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.5848113460466265, |
| "epoch": 2.0, |
| "grad_norm": 1.171875, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 0.5611, |
| "mean_token_accuracy": 0.8350322283804417, |
| "num_tokens": 41664296.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.48217232096940277, |
| "epoch": 2.016, |
| "grad_norm": 1.375, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 0.4598, |
| "mean_token_accuracy": 0.8575068596750498, |
| "num_tokens": 42001193.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.47964414674788713, |
| "epoch": 2.032, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 0.4587, |
| "mean_token_accuracy": 0.8569075904786587, |
| "num_tokens": 42340757.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.49382132217288016, |
| "epoch": 2.048, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.938e-05, |
| "loss": 0.4713, |
| "mean_token_accuracy": 0.8529856160283089, |
| "num_tokens": 42670324.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.47823069952428343, |
| "epoch": 2.064, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 0.4566, |
| "mean_token_accuracy": 0.8581176064908504, |
| "num_tokens": 43010231.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.48923970870673655, |
| "epoch": 2.08, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 0.4667, |
| "mean_token_accuracy": 0.8556209966540337, |
| "num_tokens": 43340227.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 0.546965966463089, |
| "eval_chemistry_loss": 0.7304518818855286, |
| "eval_chemistry_mean_token_accuracy": 0.8063767719268798, |
| "eval_chemistry_num_tokens": 43340227.0, |
| "eval_chemistry_runtime": 34.7458, |
| "eval_chemistry_samples_per_second": 14.39, |
| "eval_chemistry_steps_per_second": 3.598, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.49418199634552, |
| "eval_physics_loss": 0.6146922707557678, |
| "eval_physics_mean_token_accuracy": 0.8299191513061523, |
| "eval_physics_num_tokens": 43340227.0, |
| "eval_physics_runtime": 41.365, |
| "eval_physics_samples_per_second": 12.088, |
| "eval_physics_steps_per_second": 3.022, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.4770382607355714, |
| "epoch": 2.096, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 0.461, |
| "mean_token_accuracy": 0.8564476538449526, |
| "num_tokens": 43673099.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.4707287013530731, |
| "epoch": 2.112, |
| "grad_norm": 1.390625, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 0.4509, |
| "mean_token_accuracy": 0.8583695068955421, |
| "num_tokens": 44005986.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.4803559150546789, |
| "epoch": 2.128, |
| "grad_norm": 1.453125, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 0.4643, |
| "mean_token_accuracy": 0.8548851139843464, |
| "num_tokens": 44321465.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.48221837151795627, |
| "epoch": 2.144, |
| "grad_norm": 1.375, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 0.4619, |
| "mean_token_accuracy": 0.8559276178479195, |
| "num_tokens": 44661027.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.5006665755994618, |
| "epoch": 2.16, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 0.4817, |
| "mean_token_accuracy": 0.8508365921676159, |
| "num_tokens": 44994236.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.4735172586515546, |
| "epoch": 2.176, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 0.4583, |
| "mean_token_accuracy": 0.856869063898921, |
| "num_tokens": 45334280.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.47296210322529075, |
| "epoch": 2.192, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.918e-05, |
| "loss": 0.4593, |
| "mean_token_accuracy": 0.857900059223175, |
| "num_tokens": 45666954.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.47185815051198005, |
| "epoch": 2.208, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 0.452, |
| "mean_token_accuracy": 0.858792957663536, |
| "num_tokens": 45999546.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.47457288652658464, |
| "epoch": 2.224, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 0.4591, |
| "mean_token_accuracy": 0.8570701099932194, |
| "num_tokens": 46333809.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.4764716824516654, |
| "epoch": 2.24, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 0.4596, |
| "mean_token_accuracy": 0.856460390239954, |
| "num_tokens": 46673328.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 0.5586697208881378, |
| "eval_chemistry_loss": 0.7306581735610962, |
| "eval_chemistry_mean_token_accuracy": 0.8062993106842041, |
| "eval_chemistry_num_tokens": 46673328.0, |
| "eval_chemistry_runtime": 34.6938, |
| "eval_chemistry_samples_per_second": 14.412, |
| "eval_chemistry_steps_per_second": 3.603, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.5080966579914094, |
| "eval_physics_loss": 0.611297607421875, |
| "eval_physics_mean_token_accuracy": 0.830946072101593, |
| "eval_physics_num_tokens": 46673328.0, |
| "eval_physics_runtime": 41.3319, |
| "eval_physics_samples_per_second": 12.097, |
| "eval_physics_steps_per_second": 3.024, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.4922581292688847, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 1.484375, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 0.4743, |
| "mean_token_accuracy": 0.8529286984354257, |
| "num_tokens": 47001509.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.47681235875934364, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 0.4583, |
| "mean_token_accuracy": 0.8574760638177394, |
| "num_tokens": 47340142.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.4867669759318233, |
| "epoch": 2.288, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 0.4711, |
| "mean_token_accuracy": 0.853106002509594, |
| "num_tokens": 47677550.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.4855895210057497, |
| "epoch": 2.304, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 0.4641, |
| "mean_token_accuracy": 0.855696403235197, |
| "num_tokens": 47993851.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.48173528388142584, |
| "epoch": 2.32, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 0.4624, |
| "mean_token_accuracy": 0.8548369336873293, |
| "num_tokens": 48332577.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.46310502383857965, |
| "epoch": 2.336, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.898e-05, |
| "loss": 0.4466, |
| "mean_token_accuracy": 0.8590828448534011, |
| "num_tokens": 48676623.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.46570080444216727, |
| "epoch": 2.352, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 0.4477, |
| "mean_token_accuracy": 0.8588027618825436, |
| "num_tokens": 49007770.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.47828576155006886, |
| "epoch": 2.368, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 0.463, |
| "mean_token_accuracy": 0.8564012508839369, |
| "num_tokens": 49350016.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.48590457029640677, |
| "epoch": 2.384, |
| "grad_norm": 1.609375, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 0.4677, |
| "mean_token_accuracy": 0.8538792595267296, |
| "num_tokens": 49679074.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.4809177804738283, |
| "epoch": 2.4, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 0.4631, |
| "mean_token_accuracy": 0.8553439110517502, |
| "num_tokens": 50020648.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 0.5487088859081268, |
| "eval_chemistry_loss": 0.7314473986625671, |
| "eval_chemistry_mean_token_accuracy": 0.8061141548156738, |
| "eval_chemistry_num_tokens": 50020648.0, |
| "eval_chemistry_runtime": 34.6612, |
| "eval_chemistry_samples_per_second": 14.425, |
| "eval_chemistry_steps_per_second": 3.606, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.4990242130756378, |
| "eval_physics_loss": 0.6052322387695312, |
| "eval_physics_mean_token_accuracy": 0.8326480937004089, |
| "eval_physics_num_tokens": 50020648.0, |
| "eval_physics_runtime": 40.9615, |
| "eval_physics_samples_per_second": 12.207, |
| "eval_physics_steps_per_second": 3.052, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.4788668820634484, |
| "epoch": 2.416, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 0.4614, |
| "mean_token_accuracy": 0.8562805935740471, |
| "num_tokens": 50370135.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.4815561521798372, |
| "epoch": 2.432, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 0.465, |
| "mean_token_accuracy": 0.8552235502749681, |
| "num_tokens": 50695363.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.494814782217145, |
| "epoch": 2.448, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 0.4771, |
| "mean_token_accuracy": 0.851343784853816, |
| "num_tokens": 51023498.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.4887571394443512, |
| "epoch": 2.464, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 0.4677, |
| "mean_token_accuracy": 0.8553626071661711, |
| "num_tokens": 51349816.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.48406565077602864, |
| "epoch": 2.48, |
| "grad_norm": 1.5, |
| "learning_rate": 1.878e-05, |
| "loss": 0.4662, |
| "mean_token_accuracy": 0.8541974198073149, |
| "num_tokens": 51680507.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.4840725682675838, |
| "epoch": 2.496, |
| "grad_norm": 1.40625, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 0.4692, |
| "mean_token_accuracy": 0.8539839003235101, |
| "num_tokens": 52017752.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.47237055795267224, |
| "epoch": 2.512, |
| "grad_norm": 1.5390625, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 0.4547, |
| "mean_token_accuracy": 0.8578830562531948, |
| "num_tokens": 52345439.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.47300295867025854, |
| "epoch": 2.528, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 0.456, |
| "mean_token_accuracy": 0.8584080573171378, |
| "num_tokens": 52683886.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.48024273496121167, |
| "epoch": 2.544, |
| "grad_norm": 1.5546875, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 0.4609, |
| "mean_token_accuracy": 0.8557684738188982, |
| "num_tokens": 53004529.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.47996121235191824, |
| "epoch": 2.56, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 0.4592, |
| "mean_token_accuracy": 0.8556542955338955, |
| "num_tokens": 53336280.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 0.554151120185852, |
| "eval_chemistry_loss": 0.7288383841514587, |
| "eval_chemistry_mean_token_accuracy": 0.8056904196739196, |
| "eval_chemistry_num_tokens": 53336280.0, |
| "eval_chemistry_runtime": 34.6332, |
| "eval_chemistry_samples_per_second": 14.437, |
| "eval_chemistry_steps_per_second": 3.609, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.49764198398590087, |
| "eval_physics_loss": 0.6016162037849426, |
| "eval_physics_mean_token_accuracy": 0.8329321603775024, |
| "eval_physics_num_tokens": 53336280.0, |
| "eval_physics_runtime": 41.2433, |
| "eval_physics_samples_per_second": 12.123, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.46981679294258355, |
| "epoch": 2.576, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 0.4564, |
| "mean_token_accuracy": 0.8582920409739018, |
| "num_tokens": 53667647.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.47870821300894023, |
| "epoch": 2.592, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 0.4629, |
| "mean_token_accuracy": 0.8552918780595065, |
| "num_tokens": 53997614.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.47518060319125655, |
| "epoch": 2.608, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 0.4513, |
| "mean_token_accuracy": 0.8582707166671752, |
| "num_tokens": 54330195.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.48298279475420713, |
| "epoch": 2.624, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.858e-05, |
| "loss": 0.4625, |
| "mean_token_accuracy": 0.8556858822703362, |
| "num_tokens": 54664342.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.4753486838191748, |
| "epoch": 2.64, |
| "grad_norm": 1.46875, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 0.4582, |
| "mean_token_accuracy": 0.8570550225675106, |
| "num_tokens": 54993626.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.49773360043764114, |
| "epoch": 2.656, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 0.4796, |
| "mean_token_accuracy": 0.850704287365079, |
| "num_tokens": 55339962.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.48465393427759407, |
| "epoch": 2.672, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 0.4659, |
| "mean_token_accuracy": 0.8545554764568806, |
| "num_tokens": 55670567.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.4666805924847722, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 1.46875, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 0.4524, |
| "mean_token_accuracy": 0.8596824321895837, |
| "num_tokens": 56003156.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.4853338483721018, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 0.4651, |
| "mean_token_accuracy": 0.8554550468921661, |
| "num_tokens": 56342926.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.47395940907299516, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 1.5625, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 0.4578, |
| "mean_token_accuracy": 0.8566099297255277, |
| "num_tokens": 56677521.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 0.5565269174575805, |
| "eval_chemistry_loss": 0.7264519333839417, |
| "eval_chemistry_mean_token_accuracy": 0.8058576831817627, |
| "eval_chemistry_num_tokens": 56677521.0, |
| "eval_chemistry_runtime": 34.6286, |
| "eval_chemistry_samples_per_second": 14.439, |
| "eval_chemistry_steps_per_second": 3.61, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.5007153513431549, |
| "eval_physics_loss": 0.5944604277610779, |
| "eval_physics_mean_token_accuracy": 0.8342739996910096, |
| "eval_physics_num_tokens": 56677521.0, |
| "eval_physics_runtime": 41.2898, |
| "eval_physics_samples_per_second": 12.11, |
| "eval_physics_steps_per_second": 3.027, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.49342543426901103, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 0.4713, |
| "mean_token_accuracy": 0.8526182346045971, |
| "num_tokens": 57003161.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.4808831615373492, |
| "epoch": 2.752, |
| "grad_norm": 1.515625, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 0.4654, |
| "mean_token_accuracy": 0.8546452801674604, |
| "num_tokens": 57335850.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.4690231055021286, |
| "epoch": 2.768, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 0.4528, |
| "mean_token_accuracy": 0.8586602933704853, |
| "num_tokens": 57672649.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.48197978772222994, |
| "epoch": 2.784, |
| "grad_norm": 1.5, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 0.4629, |
| "mean_token_accuracy": 0.856096263974905, |
| "num_tokens": 58007431.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.5036183711141348, |
| "epoch": 2.8, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 0.4836, |
| "mean_token_accuracy": 0.8493415039032698, |
| "num_tokens": 58332730.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.462801137752831, |
| "epoch": 2.816, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 0.4433, |
| "mean_token_accuracy": 0.8603610061109066, |
| "num_tokens": 58672106.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.4801277067512274, |
| "epoch": 2.832, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 0.4664, |
| "mean_token_accuracy": 0.8545713942497969, |
| "num_tokens": 58997592.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.4840684751048684, |
| "epoch": 2.848, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 0.4668, |
| "mean_token_accuracy": 0.8554341107606888, |
| "num_tokens": 59326336.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.4763260716572404, |
| "epoch": 2.864, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 0.4575, |
| "mean_token_accuracy": 0.858063718304038, |
| "num_tokens": 59668586.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.47250860314816234, |
| "epoch": 2.88, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 0.4591, |
| "mean_token_accuracy": 0.8573315985500812, |
| "num_tokens": 59993572.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 0.5572835853099823, |
| "eval_chemistry_loss": 0.7282317876815796, |
| "eval_chemistry_mean_token_accuracy": 0.8056740365028381, |
| "eval_chemistry_num_tokens": 59993572.0, |
| "eval_chemistry_runtime": 34.6241, |
| "eval_chemistry_samples_per_second": 14.441, |
| "eval_chemistry_steps_per_second": 3.61, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.5005052845478057, |
| "eval_physics_loss": 0.5933815240859985, |
| "eval_physics_mean_token_accuracy": 0.834332944393158, |
| "eval_physics_num_tokens": 59993572.0, |
| "eval_physics_runtime": 41.2544, |
| "eval_physics_samples_per_second": 12.12, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.4770482426509261, |
| "epoch": 2.896, |
| "grad_norm": 1.5, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 0.4576, |
| "mean_token_accuracy": 0.8566633488982915, |
| "num_tokens": 60324128.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.472668781876564, |
| "epoch": 2.912, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 0.4549, |
| "mean_token_accuracy": 0.858355012908578, |
| "num_tokens": 60657399.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.4690807094797492, |
| "epoch": 2.928, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 0.4515, |
| "mean_token_accuracy": 0.8586832124739885, |
| "num_tokens": 60984711.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.4677178554236889, |
| "epoch": 2.944, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 0.4533, |
| "mean_token_accuracy": 0.858997655287385, |
| "num_tokens": 61321359.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.47983004804700613, |
| "epoch": 2.96, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 0.4607, |
| "mean_token_accuracy": 0.8569474756717682, |
| "num_tokens": 61659042.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.4755082868039608, |
| "epoch": 2.976, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 0.455, |
| "mean_token_accuracy": 0.8571343049407005, |
| "num_tokens": 61993595.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.46073452215641736, |
| "epoch": 2.992, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 0.4449, |
| "mean_token_accuracy": 0.8599927302449941, |
| "num_tokens": 62334012.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.41872623963281513, |
| "epoch": 3.008, |
| "grad_norm": 1.734375, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 0.3776, |
| "mean_token_accuracy": 0.879940589517355, |
| "num_tokens": 62660928.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.3062732330523431, |
| "epoch": 3.024, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 0.298, |
| "mean_token_accuracy": 0.9008689176291227, |
| "num_tokens": 62992670.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.314318940974772, |
| "epoch": 3.04, |
| "grad_norm": 1.890625, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 0.2908, |
| "mean_token_accuracy": 0.9024943716824054, |
| "num_tokens": 63335157.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 0.4083097501993179, |
| "eval_chemistry_loss": 0.8877620697021484, |
| "eval_chemistry_mean_token_accuracy": 0.795852870464325, |
| "eval_chemistry_num_tokens": 63335157.0, |
| "eval_chemistry_runtime": 34.5811, |
| "eval_chemistry_samples_per_second": 14.459, |
| "eval_chemistry_steps_per_second": 3.615, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.356642371058464, |
| "eval_physics_loss": 0.7040289044380188, |
| "eval_physics_mean_token_accuracy": 0.8281279788017273, |
| "eval_physics_num_tokens": 63335157.0, |
| "eval_physics_runtime": 41.2473, |
| "eval_physics_samples_per_second": 12.122, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.3101253783330321, |
| "epoch": 3.056, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.798e-05, |
| "loss": 0.2855, |
| "mean_token_accuracy": 0.9037388786673546, |
| "num_tokens": 63662314.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.3101832845248282, |
| "epoch": 3.072, |
| "grad_norm": 1.9921875, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 0.2901, |
| "mean_token_accuracy": 0.9020112063735723, |
| "num_tokens": 63997780.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.3178480817005038, |
| "epoch": 3.088, |
| "grad_norm": 1.921875, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 0.2974, |
| "mean_token_accuracy": 0.8999546930193901, |
| "num_tokens": 64325274.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.30261183911934497, |
| "epoch": 3.104, |
| "grad_norm": 1.890625, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 0.282, |
| "mean_token_accuracy": 0.9047445867210626, |
| "num_tokens": 64659683.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.29912759382277726, |
| "epoch": 3.12, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 0.2805, |
| "mean_token_accuracy": 0.9053699217736721, |
| "num_tokens": 64998212.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.31789954053238034, |
| "epoch": 3.136, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 0.2999, |
| "mean_token_accuracy": 0.8997865796089173, |
| "num_tokens": 65328436.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.3016470263712108, |
| "epoch": 3.152, |
| "grad_norm": 1.9453125, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 0.281, |
| "mean_token_accuracy": 0.9046830009669066, |
| "num_tokens": 65659346.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.313287281896919, |
| "epoch": 3.168, |
| "grad_norm": 1.8359375, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 0.2917, |
| "mean_token_accuracy": 0.9017079468816519, |
| "num_tokens": 65986382.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.3030808721669018, |
| "epoch": 3.184, |
| "grad_norm": 2.0, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 0.2839, |
| "mean_token_accuracy": 0.9039736580103636, |
| "num_tokens": 66324308.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.30991311026737095, |
| "epoch": 3.2, |
| "grad_norm": 1.8984375, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.2896, |
| "mean_token_accuracy": 0.9023264229297638, |
| "num_tokens": 66658712.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 0.3913348898887634, |
| "eval_chemistry_loss": 0.9072752594947815, |
| "eval_chemistry_mean_token_accuracy": 0.7967498378753662, |
| "eval_chemistry_num_tokens": 66658712.0, |
| "eval_chemistry_runtime": 34.6017, |
| "eval_chemistry_samples_per_second": 14.45, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.3560866928100586, |
| "eval_physics_loss": 0.7074852585792542, |
| "eval_physics_mean_token_accuracy": 0.8281491875648499, |
| "eval_physics_num_tokens": 66658712.0, |
| "eval_physics_runtime": 40.9937, |
| "eval_physics_samples_per_second": 12.197, |
| "eval_physics_steps_per_second": 3.049, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.30841554757207634, |
| "epoch": 3.216, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 0.2871, |
| "mean_token_accuracy": 0.9021381173282862, |
| "num_tokens": 66990142.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.3216819787397981, |
| "epoch": 3.232, |
| "grad_norm": 1.984375, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 0.3005, |
| "mean_token_accuracy": 0.8979394420981407, |
| "num_tokens": 67323231.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.3005070075392723, |
| "epoch": 3.248, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 0.2844, |
| "mean_token_accuracy": 0.9041368763893842, |
| "num_tokens": 67664253.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.31158956326544285, |
| "epoch": 3.2640000000000002, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 0.2891, |
| "mean_token_accuracy": 0.9025689493864775, |
| "num_tokens": 67991082.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.308028197940439, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 1.875, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 0.2913, |
| "mean_token_accuracy": 0.9015441410243511, |
| "num_tokens": 68317737.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.31388334231451154, |
| "epoch": 3.296, |
| "grad_norm": 1.890625, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 0.2952, |
| "mean_token_accuracy": 0.9001540582627058, |
| "num_tokens": 68653995.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.3085142597556114, |
| "epoch": 3.312, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 0.2916, |
| "mean_token_accuracy": 0.9023820515722036, |
| "num_tokens": 68994300.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.322191693354398, |
| "epoch": 3.328, |
| "grad_norm": 2.078125, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 0.3004, |
| "mean_token_accuracy": 0.899126460403204, |
| "num_tokens": 69323456.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.31341270450502634, |
| "epoch": 3.344, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.758e-05, |
| "loss": 0.2938, |
| "mean_token_accuracy": 0.9000593461096287, |
| "num_tokens": 69659373.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.32349290549755094, |
| "epoch": 3.36, |
| "grad_norm": 1.96875, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 0.3026, |
| "mean_token_accuracy": 0.8979135323315859, |
| "num_tokens": 69998761.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_chemistry_entropy": 0.3982364737987518, |
| "eval_chemistry_loss": 0.9002445340156555, |
| "eval_chemistry_mean_token_accuracy": 0.7955028491020203, |
| "eval_chemistry_num_tokens": 69998761.0, |
| "eval_chemistry_runtime": 34.6288, |
| "eval_chemistry_samples_per_second": 14.439, |
| "eval_chemistry_steps_per_second": 3.61, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_physics_entropy": 0.3635141521692276, |
| "eval_physics_loss": 0.6997887492179871, |
| "eval_physics_mean_token_accuracy": 0.8300124206542969, |
| "eval_physics_num_tokens": 69998761.0, |
| "eval_physics_runtime": 41.2511, |
| "eval_physics_samples_per_second": 12.121, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.3002217338420451, |
| "epoch": 3.376, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 0.2835, |
| "mean_token_accuracy": 0.9043725613504648, |
| "num_tokens": 70344663.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.3226173784583807, |
| "epoch": 3.392, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 0.3007, |
| "mean_token_accuracy": 0.8972969483584166, |
| "num_tokens": 70671470.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.31106266314163805, |
| "epoch": 3.408, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 0.2935, |
| "mean_token_accuracy": 0.9007665246725083, |
| "num_tokens": 71003257.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.3126481889747083, |
| "epoch": 3.424, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 0.2925, |
| "mean_token_accuracy": 0.9006055317819118, |
| "num_tokens": 71333695.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.31662706546485425, |
| "epoch": 3.44, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 0.3004, |
| "mean_token_accuracy": 0.8997602794319391, |
| "num_tokens": 71672438.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.31781021589413283, |
| "epoch": 3.456, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 0.3002, |
| "mean_token_accuracy": 0.8984184488654137, |
| "num_tokens": 72009302.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.3236368427053094, |
| "epoch": 3.472, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 0.3046, |
| "mean_token_accuracy": 0.8978909350931644, |
| "num_tokens": 72334603.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.31103177955374123, |
| "epoch": 3.488, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 0.2936, |
| "mean_token_accuracy": 0.9011136863380671, |
| "num_tokens": 72669209.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.324208745919168, |
| "epoch": 3.504, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 0.306, |
| "mean_token_accuracy": 0.8976800374686718, |
| "num_tokens": 73001765.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.3198518265038729, |
| "epoch": 3.52, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 0.3019, |
| "mean_token_accuracy": 0.8987439151853323, |
| "num_tokens": 73326681.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_chemistry_entropy": 0.3984639751911163, |
| "eval_chemistry_loss": 0.908758282661438, |
| "eval_chemistry_mean_token_accuracy": 0.7953772258758545, |
| "eval_chemistry_num_tokens": 73326681.0, |
| "eval_chemistry_runtime": 34.6394, |
| "eval_chemistry_samples_per_second": 14.434, |
| "eval_chemistry_steps_per_second": 3.609, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_physics_entropy": 0.36362937366962433, |
| "eval_physics_loss": 0.7028279900550842, |
| "eval_physics_mean_token_accuracy": 0.8286777052879334, |
| "eval_physics_num_tokens": 73326681.0, |
| "eval_physics_runtime": 41.2632, |
| "eval_physics_samples_per_second": 12.117, |
| "eval_physics_steps_per_second": 3.029, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.3174999676644802, |
| "epoch": 3.536, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 0.2998, |
| "mean_token_accuracy": 0.8993683304637671, |
| "num_tokens": 73658724.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.3130149595439434, |
| "epoch": 3.552, |
| "grad_norm": 1.90625, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 0.293, |
| "mean_token_accuracy": 0.9002668939530849, |
| "num_tokens": 73990446.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.3138899549841881, |
| "epoch": 3.568, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 0.2919, |
| "mean_token_accuracy": 0.8998954400420189, |
| "num_tokens": 74332932.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.3273557879962027, |
| "epoch": 3.584, |
| "grad_norm": 2.0, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 0.3092, |
| "mean_token_accuracy": 0.8954586464911699, |
| "num_tokens": 74660938.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.3111777361482382, |
| "epoch": 3.6, |
| "grad_norm": 2.125, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 0.2927, |
| "mean_token_accuracy": 0.9013317011296749, |
| "num_tokens": 75004640.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.3160340915434062, |
| "epoch": 3.616, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 0.3, |
| "mean_token_accuracy": 0.8990191236138344, |
| "num_tokens": 75344044.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.3194020535796881, |
| "epoch": 3.632, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.718e-05, |
| "loss": 0.3009, |
| "mean_token_accuracy": 0.8986658614128828, |
| "num_tokens": 75676495.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.321523059066385, |
| "epoch": 3.648, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 0.3036, |
| "mean_token_accuracy": 0.8987799856811762, |
| "num_tokens": 76005637.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.31546984519809484, |
| "epoch": 3.664, |
| "grad_norm": 1.984375, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 0.2967, |
| "mean_token_accuracy": 0.898678494617343, |
| "num_tokens": 76344141.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.32511516408994795, |
| "epoch": 3.68, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 0.3053, |
| "mean_token_accuracy": 0.897456269338727, |
| "num_tokens": 76666670.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_chemistry_entropy": 0.3999490444660187, |
| "eval_chemistry_loss": 0.905655026435852, |
| "eval_chemistry_mean_token_accuracy": 0.7958552212715149, |
| "eval_chemistry_num_tokens": 76666670.0, |
| "eval_chemistry_runtime": 34.6212, |
| "eval_chemistry_samples_per_second": 14.442, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_physics_entropy": 0.36464127004146574, |
| "eval_physics_loss": 0.700202226638794, |
| "eval_physics_mean_token_accuracy": 0.8299637427330017, |
| "eval_physics_num_tokens": 76666670.0, |
| "eval_physics_runtime": 41.2617, |
| "eval_physics_samples_per_second": 12.118, |
| "eval_physics_steps_per_second": 3.029, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.32625418398529293, |
| "epoch": 3.6959999999999997, |
| "grad_norm": 2.25, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 0.3093, |
| "mean_token_accuracy": 0.8957523014396429, |
| "num_tokens": 76997576.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.3156011073850095, |
| "epoch": 3.7119999999999997, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 0.2954, |
| "mean_token_accuracy": 0.9003329947590828, |
| "num_tokens": 77333230.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.3184890853241086, |
| "epoch": 3.7279999999999998, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 0.3026, |
| "mean_token_accuracy": 0.8980983804911375, |
| "num_tokens": 77662989.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.3237913983874023, |
| "epoch": 3.7439999999999998, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 0.3056, |
| "mean_token_accuracy": 0.897695067897439, |
| "num_tokens": 77990559.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.3197625041939318, |
| "epoch": 3.76, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 0.302, |
| "mean_token_accuracy": 0.8969969056546688, |
| "num_tokens": 78328058.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.32039122832939027, |
| "epoch": 3.776, |
| "grad_norm": 2.078125, |
| "learning_rate": 1.698e-05, |
| "loss": 0.2999, |
| "mean_token_accuracy": 0.8986205138266087, |
| "num_tokens": 78656796.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.31185208084061744, |
| "epoch": 3.792, |
| "grad_norm": 1.953125, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 0.296, |
| "mean_token_accuracy": 0.9003968022763729, |
| "num_tokens": 79000001.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.32015116736292837, |
| "epoch": 3.808, |
| "grad_norm": 1.9453125, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 0.2999, |
| "mean_token_accuracy": 0.8986766755580902, |
| "num_tokens": 79332103.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.3285049434751272, |
| "epoch": 3.824, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 0.3084, |
| "mean_token_accuracy": 0.8958335161209107, |
| "num_tokens": 79658616.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.3201352638192475, |
| "epoch": 3.84, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 0.3008, |
| "mean_token_accuracy": 0.8984422065317631, |
| "num_tokens": 79994326.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_chemistry_entropy": 0.39785524785518644, |
| "eval_chemistry_loss": 0.9046195149421692, |
| "eval_chemistry_mean_token_accuracy": 0.796384777545929, |
| "eval_chemistry_num_tokens": 79994326.0, |
| "eval_chemistry_runtime": 34.628, |
| "eval_chemistry_samples_per_second": 14.439, |
| "eval_chemistry_steps_per_second": 3.61, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_physics_entropy": 0.3649995505809784, |
| "eval_physics_loss": 0.6967271566390991, |
| "eval_physics_mean_token_accuracy": 0.8300955877304077, |
| "eval_physics_num_tokens": 79994326.0, |
| "eval_physics_runtime": 41.2661, |
| "eval_physics_samples_per_second": 12.116, |
| "eval_physics_steps_per_second": 3.029, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.3189585231244564, |
| "epoch": 3.856, |
| "grad_norm": 1.953125, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 0.2981, |
| "mean_token_accuracy": 0.8985602565109729, |
| "num_tokens": 80332596.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.3192179761826992, |
| "epoch": 3.872, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 0.3008, |
| "mean_token_accuracy": 0.8992702454328537, |
| "num_tokens": 80662674.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.32309232754632833, |
| "epoch": 3.888, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 0.3064, |
| "mean_token_accuracy": 0.8973185170441866, |
| "num_tokens": 80995198.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.3297643651254475, |
| "epoch": 3.904, |
| "grad_norm": 2.0625, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 0.3095, |
| "mean_token_accuracy": 0.8950018011033535, |
| "num_tokens": 81331457.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.32107119625434277, |
| "epoch": 3.92, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 0.3046, |
| "mean_token_accuracy": 0.8978569827973842, |
| "num_tokens": 81668354.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.31930080968886615, |
| "epoch": 3.936, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 0.3, |
| "mean_token_accuracy": 0.8981735832989216, |
| "num_tokens": 81994993.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.32174809416756034, |
| "epoch": 3.952, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 0.303, |
| "mean_token_accuracy": 0.8970499508082866, |
| "num_tokens": 82327299.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.31509073711931707, |
| "epoch": 3.968, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 0.296, |
| "mean_token_accuracy": 0.8995682530105114, |
| "num_tokens": 82653218.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.31415768414735795, |
| "epoch": 3.984, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 0.2954, |
| "mean_token_accuracy": 0.9006588537245989, |
| "num_tokens": 82997429.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.3128982843831182, |
| "epoch": 4.0, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 0.2974, |
| "mean_token_accuracy": 0.8994481891393662, |
| "num_tokens": 83328592.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_chemistry_entropy": 0.39885407757759095, |
| "eval_chemistry_loss": 0.9091149568557739, |
| "eval_chemistry_mean_token_accuracy": 0.795774167060852, |
| "eval_chemistry_num_tokens": 83328592.0, |
| "eval_chemistry_runtime": 34.6043, |
| "eval_chemistry_samples_per_second": 14.449, |
| "eval_chemistry_steps_per_second": 3.612, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_physics_entropy": 0.36530122339725496, |
| "eval_physics_loss": 0.6973084211349487, |
| "eval_physics_mean_token_accuracy": 0.8300330719947815, |
| "eval_physics_num_tokens": 83328592.0, |
| "eval_physics_runtime": 40.9427, |
| "eval_physics_samples_per_second": 12.212, |
| "eval_physics_steps_per_second": 3.053, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.21543310643173755, |
| "epoch": 4.016, |
| "grad_norm": 2.34375, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 0.162, |
| "mean_token_accuracy": 0.9484033465385437, |
| "num_tokens": 83655133.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.19175719134509564, |
| "epoch": 4.032, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 0.1519, |
| "mean_token_accuracy": 0.9507987700402737, |
| "num_tokens": 83995704.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.19031685492955147, |
| "epoch": 4.048, |
| "grad_norm": 1.921875, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 0.1504, |
| "mean_token_accuracy": 0.9515091743320226, |
| "num_tokens": 84323657.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.18541114097461103, |
| "epoch": 4.064, |
| "grad_norm": 1.921875, |
| "learning_rate": 1.658e-05, |
| "loss": 0.1469, |
| "mean_token_accuracy": 0.9520056396722794, |
| "num_tokens": 84662055.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.18697019712999463, |
| "epoch": 4.08, |
| "grad_norm": 1.984375, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 0.1469, |
| "mean_token_accuracy": 0.9523114804178476, |
| "num_tokens": 84990884.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.18290056772530078, |
| "epoch": 4.096, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 0.1474, |
| "mean_token_accuracy": 0.9523672234266997, |
| "num_tokens": 85334346.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.18544434658251702, |
| "epoch": 4.112, |
| "grad_norm": 2.125, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 0.1467, |
| "mean_token_accuracy": 0.9518146891146898, |
| "num_tokens": 85654431.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.18685742230154573, |
| "epoch": 4.128, |
| "grad_norm": 2.125, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 0.148, |
| "mean_token_accuracy": 0.9512113470584154, |
| "num_tokens": 85984868.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.19103994900360705, |
| "epoch": 4.144, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 0.1533, |
| "mean_token_accuracy": 0.9495586734265089, |
| "num_tokens": 86313610.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.1833164133131504, |
| "epoch": 4.16, |
| "grad_norm": 1.96875, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 0.1491, |
| "mean_token_accuracy": 0.950814313814044, |
| "num_tokens": 86648308.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_chemistry_entropy": 0.3191299349069595, |
| "eval_chemistry_loss": 1.0713955163955688, |
| "eval_chemistry_mean_token_accuracy": 0.7893559455871582, |
| "eval_chemistry_num_tokens": 86648308.0, |
| "eval_chemistry_runtime": 34.5895, |
| "eval_chemistry_samples_per_second": 14.455, |
| "eval_chemistry_steps_per_second": 3.614, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_physics_entropy": 0.26984936237335205, |
| "eval_physics_loss": 0.8553917407989502, |
| "eval_physics_mean_token_accuracy": 0.8235324311256409, |
| "eval_physics_num_tokens": 86648308.0, |
| "eval_physics_runtime": 41.2124, |
| "eval_physics_samples_per_second": 12.132, |
| "eval_physics_steps_per_second": 3.033, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.1851897793356329, |
| "epoch": 4.176, |
| "grad_norm": 2.125, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 0.1499, |
| "mean_token_accuracy": 0.9512480426579714, |
| "num_tokens": 86979720.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.1831663493067026, |
| "epoch": 4.192, |
| "grad_norm": 1.984375, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 0.1477, |
| "mean_token_accuracy": 0.9515604998916387, |
| "num_tokens": 87322940.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.18741083294153213, |
| "epoch": 4.208, |
| "grad_norm": 2.0, |
| "learning_rate": 1.638e-05, |
| "loss": 0.149, |
| "mean_token_accuracy": 0.9508396230638028, |
| "num_tokens": 87654476.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.19511286839842795, |
| "epoch": 4.224, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 0.1587, |
| "mean_token_accuracy": 0.9478027503937483, |
| "num_tokens": 87986180.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.18864089637063444, |
| "epoch": 4.24, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 0.153, |
| "mean_token_accuracy": 0.9495376665145159, |
| "num_tokens": 88322431.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.18742439663037658, |
| "epoch": 4.256, |
| "grad_norm": 2.0625, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 0.1525, |
| "mean_token_accuracy": 0.9499348193407059, |
| "num_tokens": 88650585.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.19175748783163726, |
| "epoch": 4.272, |
| "grad_norm": 2.390625, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 0.1569, |
| "mean_token_accuracy": 0.9484076362103224, |
| "num_tokens": 88983258.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.1896573031321168, |
| "epoch": 4.288, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 0.1533, |
| "mean_token_accuracy": 0.949416034668684, |
| "num_tokens": 89306176.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.1905643475241959, |
| "epoch": 4.304, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 0.1542, |
| "mean_token_accuracy": 0.949146980792284, |
| "num_tokens": 89635604.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.19005935504101218, |
| "epoch": 4.32, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 0.1539, |
| "mean_token_accuracy": 0.9488909468054771, |
| "num_tokens": 89961895.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_chemistry_entropy": 0.3175645558834076, |
| "eval_chemistry_loss": 1.0784482955932617, |
| "eval_chemistry_mean_token_accuracy": 0.789765251159668, |
| "eval_chemistry_num_tokens": 89961895.0, |
| "eval_chemistry_runtime": 34.6063, |
| "eval_chemistry_samples_per_second": 14.448, |
| "eval_chemistry_steps_per_second": 3.612, |
| "step": 2700 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_physics_entropy": 0.2694563043117523, |
| "eval_physics_loss": 0.8580674529075623, |
| "eval_physics_mean_token_accuracy": 0.8243379302024841, |
| "eval_physics_num_tokens": 89961895.0, |
| "eval_physics_runtime": 41.2089, |
| "eval_physics_samples_per_second": 12.133, |
| "eval_physics_steps_per_second": 3.033, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.18589448984712362, |
| "epoch": 4.336, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 0.1502, |
| "mean_token_accuracy": 0.9505696080625057, |
| "num_tokens": 90293218.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.18639519005082547, |
| "epoch": 4.352, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.618e-05, |
| "loss": 0.1536, |
| "mean_token_accuracy": 0.9487312618643046, |
| "num_tokens": 90631006.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.19433019501157106, |
| "epoch": 4.368, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 0.1593, |
| "mean_token_accuracy": 0.9474144730716944, |
| "num_tokens": 90951742.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.1920853616669774, |
| "epoch": 4.384, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 0.1574, |
| "mean_token_accuracy": 0.9479308255016804, |
| "num_tokens": 91290391.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.18902287022210656, |
| "epoch": 4.4, |
| "grad_norm": 2.015625, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 0.1541, |
| "mean_token_accuracy": 0.9490096058696509, |
| "num_tokens": 91623212.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.1853318156208843, |
| "epoch": 4.416, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 0.1516, |
| "mean_token_accuracy": 0.9501545939594507, |
| "num_tokens": 91963726.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.18632075604982673, |
| "epoch": 4.432, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 0.1533, |
| "mean_token_accuracy": 0.9498950578272343, |
| "num_tokens": 92295130.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.19441989213228225, |
| "epoch": 4.448, |
| "grad_norm": 2.1875, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 0.1587, |
| "mean_token_accuracy": 0.9473429799079895, |
| "num_tokens": 92619914.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.18935320070013403, |
| "epoch": 4.464, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 0.1563, |
| "mean_token_accuracy": 0.9483810655772686, |
| "num_tokens": 92954334.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.1866440698504448, |
| "epoch": 4.48, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 0.1545, |
| "mean_token_accuracy": 0.9490117151290178, |
| "num_tokens": 93288025.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_chemistry_entropy": 0.32049827003479003, |
| "eval_chemistry_loss": 1.0742449760437012, |
| "eval_chemistry_mean_token_accuracy": 0.7895661110877991, |
| "eval_chemistry_num_tokens": 93288025.0, |
| "eval_chemistry_runtime": 34.6079, |
| "eval_chemistry_samples_per_second": 14.448, |
| "eval_chemistry_steps_per_second": 3.612, |
| "step": 2800 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_physics_entropy": 0.272767622590065, |
| "eval_physics_loss": 0.8561501502990723, |
| "eval_physics_mean_token_accuracy": 0.8236871433258056, |
| "eval_physics_num_tokens": 93288025.0, |
| "eval_physics_runtime": 41.2289, |
| "eval_physics_samples_per_second": 12.127, |
| "eval_physics_steps_per_second": 3.032, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.19370509777218103, |
| "epoch": 4.496, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 0.1597, |
| "mean_token_accuracy": 0.9468937519937753, |
| "num_tokens": 93621454.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.19159375540912152, |
| "epoch": 4.5120000000000005, |
| "grad_norm": 2.0625, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 0.1566, |
| "mean_token_accuracy": 0.9483846995979548, |
| "num_tokens": 93957847.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.19177836943417786, |
| "epoch": 4.5280000000000005, |
| "grad_norm": 2.078125, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 0.1595, |
| "mean_token_accuracy": 0.9476821962743998, |
| "num_tokens": 94287674.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.19004529607482254, |
| "epoch": 4.5440000000000005, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 0.1565, |
| "mean_token_accuracy": 0.9476745925843716, |
| "num_tokens": 94613456.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.19111841106787325, |
| "epoch": 4.5600000000000005, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 0.1586, |
| "mean_token_accuracy": 0.9473724879324437, |
| "num_tokens": 94941684.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.19352899109944702, |
| "epoch": 4.576, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 0.1578, |
| "mean_token_accuracy": 0.9478868711739779, |
| "num_tokens": 95278037.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.1918213707394898, |
| "epoch": 4.592, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 0.1574, |
| "mean_token_accuracy": 0.9480335034430027, |
| "num_tokens": 95611182.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.19304955611005425, |
| "epoch": 4.608, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 0.16, |
| "mean_token_accuracy": 0.9472247928380966, |
| "num_tokens": 95952986.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.18729796851985156, |
| "epoch": 4.624, |
| "grad_norm": 2.015625, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 0.1564, |
| "mean_token_accuracy": 0.9481069557368755, |
| "num_tokens": 96297175.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.1974953768774867, |
| "epoch": 4.64, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.578e-05, |
| "loss": 0.1634, |
| "mean_token_accuracy": 0.9463304281234741, |
| "num_tokens": 96630392.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_chemistry_entropy": 0.319516619682312, |
| "eval_chemistry_loss": 1.0775052309036255, |
| "eval_chemistry_mean_token_accuracy": 0.7886076564788819, |
| "eval_chemistry_num_tokens": 96630392.0, |
| "eval_chemistry_runtime": 34.5969, |
| "eval_chemistry_samples_per_second": 14.452, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 2900 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_physics_entropy": 0.27287652349472047, |
| "eval_physics_loss": 0.8562257289886475, |
| "eval_physics_mean_token_accuracy": 0.8237788276672363, |
| "eval_physics_num_tokens": 96630392.0, |
| "eval_physics_runtime": 41.247, |
| "eval_physics_samples_per_second": 12.122, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.18674168265424668, |
| "epoch": 4.656, |
| "grad_norm": 2.078125, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 0.1539, |
| "mean_token_accuracy": 0.9486437644809484, |
| "num_tokens": 96978935.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.1915013926103711, |
| "epoch": 4.672, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 0.1604, |
| "mean_token_accuracy": 0.947072671726346, |
| "num_tokens": 97316431.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.19037255314178764, |
| "epoch": 4.688, |
| "grad_norm": 2.125, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 0.158, |
| "mean_token_accuracy": 0.9471198979765176, |
| "num_tokens": 97665534.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.19228512230329214, |
| "epoch": 4.704, |
| "grad_norm": 2.265625, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 0.1588, |
| "mean_token_accuracy": 0.9474438637495041, |
| "num_tokens": 98004052.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.18908405187539756, |
| "epoch": 4.72, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 0.1562, |
| "mean_token_accuracy": 0.9483963634818793, |
| "num_tokens": 98325785.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.19748488427139818, |
| "epoch": 4.736, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 0.161, |
| "mean_token_accuracy": 0.9457533366978168, |
| "num_tokens": 98649969.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.18734354763291777, |
| "epoch": 4.752, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 0.1563, |
| "mean_token_accuracy": 0.947586378827691, |
| "num_tokens": 98987145.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.19386245338246227, |
| "epoch": 4.768, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 0.1603, |
| "mean_token_accuracy": 0.9473159741610289, |
| "num_tokens": 99309770.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.19166214000433685, |
| "epoch": 4.784, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 0.1588, |
| "mean_token_accuracy": 0.9476859755814075, |
| "num_tokens": 99643583.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.19860149454325438, |
| "epoch": 4.8, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 0.1616, |
| "mean_token_accuracy": 0.946180134639144, |
| "num_tokens": 99972302.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_chemistry_entropy": 0.3164873479604721, |
| "eval_chemistry_loss": 1.07830011844635, |
| "eval_chemistry_mean_token_accuracy": 0.7890521488189697, |
| "eval_chemistry_num_tokens": 99972302.0, |
| "eval_chemistry_runtime": 34.601, |
| "eval_chemistry_samples_per_second": 14.45, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_physics_entropy": 0.27059548449516296, |
| "eval_physics_loss": 0.8571956753730774, |
| "eval_physics_mean_token_accuracy": 0.824473412513733, |
| "eval_physics_num_tokens": 99972302.0, |
| "eval_physics_runtime": 40.9799, |
| "eval_physics_samples_per_second": 12.201, |
| "eval_physics_steps_per_second": 3.05, |
| "step": 3000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 16, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 7.43836977664732e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
